An inverse Wishart prior [IW(6,4)] was set on the error covariance matrix ∑ and a matrix normal prior [N(0.02,∑)] on the vector of parameters, according to recommendations of the authors. Method ‘expected’ was applied, which results in the use of expected genotype counts (∼dosages) in the analyses. The output file contains a log10 Bayes Factor (BF) per genetic variant, which was extracted for the purpose of this study.
